1. Field of the Invention
The present invention relates to workload management in computer systems. More specifically, the present invention relates to the live migration of virtual machines.
2. Background of the Related Art
In a cloud computing environment, a management system performs constant monitoring of numerous compute nodes that make up the cloud. Accordingly, the management system may take steps to balance the load among the individual compute nodes, and may deploy workloads to compute nodes, such as servers, that are able to perform appropriately.
In various implementations, the management system is the centralized point for workload distribution and has access to information about the condition of all servers and all workloads. The management system collects this information and uses it to determine how to move and rearrange workloads to accomplish one or more operational objective. The management system, such as a management server, may be limited as to what information it can collect about the individual servers or hosts. In some systems, server operating conditions are exposed to the management server so that the management server can make a more informed decision as to which host should run a given workload. However, as more information is shared with the management system, the management system may require hardware and software upgrades in order to be able to manage the larger infrastructure.
Live migration is a technology that enables movement of a virtual machine, and a guest application running in the virtual machine, from one physical host to another physical host without having to stop the guest application. The memory and execution state of the running guest application is copied over the network from one host to another to achieve the migration.
Live migration has significant advantages in the virtualized and cloud environment. For example, live migration may be used to manage resources in a cloud computing environment, where a single cloud computing provider may have thousands of guests running in their data centers. In order to save energy, reduce costs and balance the load on the servers, the provider will move the guests among the servers using live migration without disrupting their customers' guest applications. In a situation where resources, such as a processor or memory, are constrained on one physical server in the cloud, a guest application can be migrated to a less-heavily loaded server within a migration domain.
Another significant advantage of live migration is the fact that it facilitates proactive maintenance. If an imminent failure is suspected, the guest applications can be migrated off of the suspected hardware and continue running without disruption of service. Once the potential problem has been resolved, the suspected hardware may be made available to run various applications.